On Thu, Apr 07, 2016 at 12:24:58PM +0800, Liming Wang wrote:
On Thu, Apr 07, 2016 at 10:32:41AM +0800, Leo Yan wrote:
On Thu, Apr 07, 2016 at 10:04:42AM +0800, Leo Yan wrote:
On Thu, Apr 07, 2016 at 12:45:56AM +0800, Liming Wang wrote:
On Tue, Apr 05, 2016 at 03:33:38PM -0700, Jorge Ramirez-Ortiz wrote:
On 04/05/2016 09:51 AM, Liming Wang wrote:
I used the latest Image and hi6220-hikey.dtb from . And boot log is attached. I don't know whether there is anything block the kernel to boot.
Starting kernel ...
[ 0.000000] Initializing cgroup subsys cpuset [ 0.000000] Initializing cgroup subsys cpu [ 0.000000] Initializing cgroup subsys cpuacct [ 0.000000] Linux version 3.18.0-linaro-hikey (buildslave@x86-64-08) (gcc version 5.1.1 20150608 (Linaro GCC 5.1-2015.08) ) #1 SMP PREEMPT Thu Feb 25 07:17:29 UTC 2016
BTW, I found you are using Linux kernel 3.18. So looks like you are
Yes, I got the kernel from linaro:: https://builds.96boards.org/snapshots/hikey/linaro/debian/latest/
I think it's the latest.:)
Thanks for confirmation. I'm start to suspect the memory region passed from uboot is not same with UEFI. So when reserve memory nodes in mailbox's dts patch, it's out of the range of memory region which initalized in Kernel and finally introduce panic.
So for panic issue, UEFI will have no such issue, due UEFI will pass memory regions from EFI stub, so this panic will be only related with uboot's booting.
I will try at my side with uboot's booting as well.
using old kernel version with a new dtb file. I'm not sure if there have mismatching issue (such like PSCI and CPU node). Now we have moved Linux kernel to 4.1:
OK, I will try this kernel. Does 4.1 kernel support all features that 3.18 has supported?
Yes, please try 4.1, Let's work on the same code base firstly.
I think 4.1 kernel at least can resolve the issue for SMP booting. Please note, as you said in another email, 96baord's 15.05 version uses spin-table method rather than psci. So I think this is the reason for SMP booting's failure.
But like I said above, suppose the kernel version is not the root cause for memory region issue. If so please go back to report at here and it will be bug to be fixed.
BTW, is there plan to move 4.4 kernel?
I think so. I'd like leave this question to John or Amit to answer.
Thanks, Leo Yan